Performance Rating
AMD Radeon Instinct MI300X outperforms AMD Radeon Pro W6500M by 3,477.95% in the overall GPU ARK performance rating
AMD Radeon Pro W6500M
The AMD Radeon Instinct MI300X dramatically outperforms the AMD Radeon Pro W6500M, leading by 3478% in the overall GPI performance index. For machine learning, the AMD Radeon Instinct MI300X has the edge in FP16 performance (+7723%) and VRAM (192 vs 4 GB). For gaming, the AMD Radeon Instinct MI300X delivers 1856% more FP32 throughput, and the AMD Radeon Pro W6500M supports hardware ray tracing. In performance per watt, the AMD Radeon Pro W6500M is 53% more efficient at 25W TDP vs 750W for the AMD Radeon Instinct MI300X.
Contents:
Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Benchmarks AdditionalMemory
Memory Size
+4,700%
192
ГБ
Memory Type
Memory Bandwidth
10.3 TB/s
Memory Bus Width
ML Performance
FP16 (Half Precision)
+7,723%
653.7
TFLOPS
BF16 (Brain Float)
TF32 (TensorFloat)
Compute Power
FP32 (Single Precision)
+1,856%
81.72
TFLOPS
FP64 (Double Precision)
+31,198%
81.72
TFLOPS
CUDA Cores
+1,800%
19,456
RT Cores
Architecture & Compatibility
GPU Architecture
SM (Streaming Multiprocessor)
PCIe Version
ML Software Support
CUDA Version
Clocks & Performance
Base Clock
1,000
Boost Clock
+3%
2,100
Memory Clock
+26%
2,525
Power Consumption
Recommended PSU
Power Connector
TDP/TGP
-97%
25 W
Rendering
Texture Units (TMU)
+1,800%
1,216
ROP
L2 Cache
16 MB
Benchmarks
MLPerf, llama2-70b-99.9 (UNSET)
MLPerf, llama2-70b-99.9 (fp16)
MLPerf, llama2-70b-99.9 (fp8)
MLPerf, llama3.1-405b (UNSET)
MLPerf, llama3.1-405b (fp16)
llama.cpp, llama-2-7b-Q4_0
MLPerf, mixtral-8x7b (fp8)
Additional
Slots
Release Date
Display Outputs
Renting is cheaper than buying
H100 80Gb SXM (80GB)
NVIDIA
Флагманская GPU на архитектуре Hopper с 80 ГБ памяти HBM3 в форм-факторе SXM. Создана для обучения и инференса больших языковых моделей, HPC задач. Поддерживает FP64, FP32, TF32, FP16, BF16, FP8 и INT8 вычисления с Tensor Cores 4-го поколения.
from 299.11 RUB/hour
